Vol. 23, No. 1, 2021 DOI: 10.37190/ABB-01753-2020-02
The electromyographic activity characteristics of the gluteus medius muscle
before and after total hip arthroplasty
KONRAD KOPEĆ1, PRZEMYSŁAW BEREZA1*, GRZEGORZ SOBOTA2, GRZEGORZ HAJDUK3, DAMIAN KUSZ1
1 Department of Orthopedics and Traumatology, Faculty of Medical Sciences in Katowice, Medical University of Silesia, Katowice, Poland.
2 Institute of Sport, Department of Human Motor Behaviour, The Jerzy Kukuczka Academy of Physical Education, Katowice, Poland.
3 Department of Functional Diagnostics and Sports Medicine, The Jerzy Kukuczka Academy of Physical Education, Katowice, Poland.
Purpose: The clinical outcomes of total hip arthroplasty are influenced by the correct muscle function that determines good, long- term and proper function of the artificial joint. The aim of the study was to analyze the electromyographic activity of the gluteus medius muscle in patients with hip osteoarthritis and after arthroplasty in various static weight bearing conditions, both on the affected and con- tralateral side. Methods: The prospective study involved 70 patients qualified for hip replacement. Patients underwent a surface electro- myography of the gluteus medius muscle which involved the Trendelenburg test. The normalized results were obtained for both hips, preoperatively and 6 months after arthroplasty. Results: The only muscle activity differences were found at a full load condition of lower limb. In the preoperative assessment, the activity of the gluteus medius muscle was greater on the side qualified for surgery. After arthro- plasty and the rehabilitation period, the muscle activity on the operated side decreased and significantly increased on the contralateral side. Detailed analysis of the contralateral side revealed relationship with osteoarthritis. Previous hip arthroplasty of that side resulted in lower muscle activity, similar to fully functional joints. Conclusion: The activity characteristics of the gluteus medius muscle vary de- pending on the condition of the joint, and the characteristics change as a result of the surgical procedure performed on both the operated and contralateral sides. These dependencies should be taken into account in the rehabilitation process, especially at the side opposite to the operated one.
Key words: osteoarthritis, total hip replacement, gluteus medius activity, hip joint
1. Introduction
Hip osteoarthritis (OA) significantly disturbs the complicated biomechanical system of the hip joint.
The goal of surgical treatment is the hip joint recon- struction using implants [11], [17]. Biomechanical re- construction of the musculoskeletal system is signifi- cantly influenced by the correct muscle function that determines good, long-term and correct function of the artificial joint. The subject of the study was to assess
the function of the gluteus medius (GM) muscle, one of the most important muscle conditioning the proper biomechanics of the hip joint [3], [20]. The Trendel- enburg test is a functional test that was included in the study protocol. It measures the frontal plane tilt angle of the pelvis during single leg stance. A negative Trendel- enburg test reflects desirable hip abductor muscle function and stable gait, whereas a positive Trendel- enburg test is associated with hip abductor dysfunc- tion and gait disturbances [8]. This test is now used worldwide as a method for evaluating hip abductor
______________________________
* Corresponding author: Przemysław Bereza, ul. Ziołowa 45/47, 40-635 Katowice, Poland. E-mail address: pbereza@sum.edu.pl Received: November 17th, 2020
Accepted for publication: February 11th, 2021
function after total hip arthroplasty (THA) [19]. In addi- tion, electromyographic examination has been used in the analysis of hip abduction muscle sufficiency be- cause it made it possible to measure timing and rela- tive intensity of muscle activation. The electromyog- raphy (EMG) signals of GM muscle were analyzed in patients with hip OA and in patients after total hip arthroplasty [10]. Undoubtedly, disease of one hip joint has a significant impact on the entire musculoskeletal system, especially on the contralateral joint [9]. Most of the available publications concern the functional evaluation of the operated hip joint both before and after THA. The subject of this study was also the op- posite joint, regardless of its clinical condition [10], [12], [13]. Assuming the proper biomechanical recon- struction of the hip OA, we expect an improvement in muscle function of both hips. The aim of this study was to analyze the electromyographic activity of the GM muscle in patients with hip OA before and after THA in various static weight bearing conditions on both affected and contralateral hips.
2. Materials and methods
Participants
70 patients were qualified to this prospective study.
The patients were operated in the Department of Or- thopedics and Traumatology of the Medical University of Silesia in Katowice, Faculty of Medical Sciences in Katowice. The research were conducted between 2015–2019. Due to different technical and medical reason only 64 patients finished evaluation (charac- teristic of subjects in Table 1).
The main inclusion criterion was primary hip OA, and the exclusion criteria included: 1) secondary hip OA, excluding the low-grade developmental dysplasia of the hip joint (Crowe > 1) [23]; 2) performed surgi- cal treatment of the hip prior to qualification for ar- throplasty (arthroscopy, removal of the femoroacetabular impingement, osteotomy and others); 3) shortening of
a leg qualified for arthroplasty > 3 cm; 4) history of severe leg injuries followed by surgical treatment or prolonged immobilization; 5) neurological disorders, including paresis and radiculopathies; 6) malforma- tions of the lower limbs; 7) performed arthroplasty of another lower limb joint (knee or ankle joint); 8) per- formed arthrodesis or lower limb osteotomy; 9) THA performed in another department; 10) time from ar- throplasty less than 6 months or longer than 8 months;
11) rehabilitation conducted contrary to the protocol requested in the study; 12) lack of patient consent for participation in the study.
Because of the polyarticular and heterogenic na- ture of OA, the patients qualified for the study con- stituted a heterogeneous group. Some of the patients qualified for THA were also diagnosed with contralat- eral hip OA and qualified for conservative treatment or contralateral THA at a later time. Another group of patients included individuals with unilateral hip OA and normal contralateral hip. The third group con- sisted of patients qualified for hip replacement after performed previously contralateral THA. Due to the aforementioned heterogeneity of patients, it was de- cided to treat each individual joint as a separate entity and subject of the analysis. The study did not take sided dominance into account. Individual 128 hips were assigned to the following groups:
• O – hip OA qualified for arthroplasty (examina- tion before, N = 38) and after performed THA (ex- amination after, N = 64),
• C – contralateral joint to operated hip (before, N = 38; after, N = 64),
and in the detailed condition of the hip:
• CH – healthy hip, contralateral to operated hip (be- fore, N = 14; after, N = 28),
• CE – previously performed THA, contralateral to operated hip (before, N = 12; after, N = 14),
• CC – hip joint affected by OA, contralateral to oper- ated hip (before, N = 12; after, N = 22).
The subjects were informed about the study design and signed informed consent to the study. The research was approved by the Bioethics Committee of the Medi- cal University of Silesia in Katowice.
Table 1. The subjects’ body mass and age separate for evaluation time (before and after surgery) and sex Body mass [kg] (mean ± STD) Age [years] (mean ± STD)
Period N
W M Total W M Total
Before 38
(24 W, 14 M) 64.3 ± 11.2 85.1 ± 10.7 73.1 ± 14.4 62.8 ± 14.8 55.2 ± 13.6 59.5 ± 14.5
After 64
(40 W, 24 M) 68.9 ± 10.8 82.3 ± 13.1 74.5 ± 13.2 63.4 ± 14.2 56.1 ± 12.8 61.2 ± 13.3 N – group size, W – women, M – men, STD – standard deviation.
Experimental procedures and instruments. The study used surface EMG in a bipolar system (two active receiving electrodes) with a TeleMyo 400 recorder (Noraxon, USA). The recorder is a portable four- channel recorder with a 1000 Hz signal sampling fre- quency. Dedicated MyoResearch XP Master Edition software (Noraxon, USA) was used to support the recorder, signal processing and data archiving. The prin- ciples of patient preparation for EMG, electrode place- ment and the entire study were in accordance with the procedure SENIAM (Surface Electromyography for the Non-Invasive Assessment of Muscles) and ISEK (International Society of Electromyography and Kine- siology) [2], [18]. Active electrodes were placed for the GM at 1/2 distance measured between the iliac crest and the greater femoral trochanter along the muscle course, at a distance of about 15–20 mm.
The electromyographic activity of the GM of both hips was measured on the test stand for normalizing the results to the maximum voluntary isometric con- traction (MVIC) [4]. The measuring stand consisted of an examination bed with a movable resistor. The MVIC test was performed lying on side. An adjustable ab- ductor resistor was placed at the height of the patient’s fibula head, and the resistance was set at a height equal to the width of the pelvis. Patients sought to perform hip abduction (repeated 5 times) with their maximum commitment to evaluate the MVIC. Next, the Tren- delenburg test were performed and muscle activity were measured for a support limb (loaded with around 80% of body weight – Load A), and for a lifted leg (loaded by leg weight, around 20% of body weight – Load B). EMG recording was also performed after changing the weight bearing limbs. The mean values of normalized bioelectrical activity of the GM (mGM) were submitted for statistical analysis.
Patients were then operated. The cementless press-fit arthroplasties were performed by one team of sur- geons. From the first day after operation patients were encouraged to get out of bed, walk with frame and then with crutches. Physical rehabilitation lasted up to 7–10 days after surgery. During early postop- erative home rehabilitation patients implemented a rehabilitation program and walked with two crutches non weight bearing for the next 4 weeks. For the following 4 weeks patients walked with partial weight bearing with one crutch held in the opposite hand to the operated side. All patients underwent hospital rehabilitation after 12 weeks from operation, which lasted 3 weeks. Within about 6 months after THA patients were referred for the next surface EMG of GM of both hips, according to the previous scheme.
Statistical analysis. Two-factor analysis of variance was used to determine the effect of hip condition and load value of lower limb onto muscle activity of GM.
A significance level of α = 0.05 was assumed for all tests.
Due to non-compliance with the distribution nor- mality condition by all compared groups and frequent right-skewed distribution, logarithmic transformation of the data was performed. As a result, the normal distri- bution confirmed by the Shapiro–Wilk test and homo- geneity of variance confirmed by the Leven test was obtained [22].
3. Results
The median values (and first-third quartiles) of muscle activity (calculated as % MVIC) due to the hip condition, limb load and period of observation are presented in Table 2.
The first step was to analyze the differences be- tween bioelectric activity within the operated “O” and contralateral joints “C” according to the load limb con- dition before (Fig. 1) and after THA (Fig. 2).
The analysis of results before surgery indicated significant higher level of muscle activity for tests with full weight bearing (F (1.44) = 13.8; p < 0.005), which could be expected. However, it was surprising that for the group of contralateral joints there was no significant differentiation between the load condi- tions. Also, with the low limb load (only leg weight,
“Load B”), similar activity level was recorded for the group O and C. The difference between groups dur- ing full load (body weight, “Load A”) pointed out to higher muscle activity of side qualified to the hip surgery (group O), but was not confirmed statisti- cally.
The results after surgery (Fig. 2) showed that the load factor (A/B load) significantly affects the higher values of muscle activity while loading the limb of the whole-body weight (F (1.120) = 79.56; p < 0.0001), as before treatment. Also, low load condition (“Load B”) showed results similar as in the first evaluation. We observed opposite characteristic during full load con- ditions, because of decreased muscle activity on oper- ated side (from 40 to 31%, group O) and increased on non-operated side (from 25 to 41%, group C).
Due to the fact that joints opposite to those affected by OA were a range of clinical state, from healthy joints, through OA, to cases after arthroplasty, it was decided to deepen this analysis, considering the con- dition of the contralateral joint in detail.
The mean values before surgery indicate that joints qualified for surgery (O) as well as opposing joints affected by OA (CC) present the highest val- ues in full weight bearing tests (Load A, Fig. 3). The
cases of healthy joints and those previously operated showed lower activity (CE and CH, Load A, Fig. 3), slightly higher than in the case of low weight bear- ing.
Table 2. Median value and Quartiles (first – third) of mean muscle activity (% MVIC) in the Trendelenburg test separate for load condition (A and B), hip condition (O and C side; O, CC, CE and CH group) and evaluation time (before and after surgery)
Load Side Period N Median
(Q1–Q3) Group Period N Median
(Q1–Q3) Before 38 36.2 (21–54)
O After 64 28.2 (23–40)
Before 12 28.9 (25–38)
CC After 22 43.8 (35–73)
Before 38 26.3
(16–35)
Before 12 17.6 (14–26)
CE After 14 23.0 (18–35)
Before 14 29.4 (12–39) Test A
C
After 64 37.7
(23–52) CH
After 28 37.8 (19–52) Before 38 14.6 (7–19)
O After 64 11.4 (6–18)
Before 12 12.7 (10–17)
CC After 22 13.0 (8–20)
Before 38 12.4
(6–16)
Before 12 9.6 (5–21)
CE After 14 12.3 (4–25)
Before 14 12.4 (6–16) Test B
C
After 64 12.3
(7–18) CH
After 28 10.5 (7–16)
N – group size, Q1 – first quartile, Q3 – third quartile.
Fig. 1. Mean normalized GM activity [%MVIC] of the operated (O) and contralateral (C) sides for Load A and B condition before arthroplasty
Fig. 2. Mean normalized GM activity [%MVIC] of the operated (O) and contralateral (C) sides for Load A and B condition after arthroplasty
Fig. 3. Mean normalized GM activity [%MVIC] for detailed classification of hip joints for Load A and B condition before arthroplasty
The lack of differentiation based on the type of load observed previously (Fig. 1) for whole group C, now has better view of intra group variability (Fig. 3). The analysis of variance indicates only the significant factor of the type of load (F (1.40) = 6.69; p < 0.05), while the post hoc analysis has only one significant difference – a higher mean GM muscle activity for operated side (group O, p = 0.015).
The significant effect of leg load condition at sec- ond evaluation pointed on higher muscle activity at
“Load A” (F (1.116) = 59.27; p < 0.0001), but post hoc tests showed no differentiation only in the CE group due to the size of the load (Fig. 4).
However, it turned out that the assignment of hips to detailed categories significantly differentiates the results obtained (F (3.116) = 3.01; p = 0.033), and a detailed analysis indicates significantly higher val- ues of muscle activity in the CC group compared to the CE group ( p = 0.048). For a full weight bearing the joints left for conservative treatment or subse- quent intervention obtains the highest activity, while operated joints (at this treatment or previously) pres- ent a level similar to healthy joints. For “Load B”
conditions no differences were found regardless of hip condition.
4. Discussion
An increased muscle activity of the affected hip was noticed in static tests with a full weight bearing condition in the initial examination, although not sta- tistically confirmed. Under low weight bearing condi- tions a symmetrical distribution of muscle activity was noticed in both hips. The arthroplasty procedure ap- proximates the characteristics of the mean bioelectri- cal activity of both, the operated and contralateral hip joints in static test conditions and emphasizes the di- versity in the muscle tension due to the state of weight bearing. After arthroplasty, the GM muscle opposite to the operated hip, affected by OA, revealed a higher bioelectric activity.
Osteoarthritis is often a polyarticular disease. Pa- tients qualified for arthroplasty have advanced hip dys- function what could result from long conservative treatment. This dysfunction affects the structures of the entire joint and muscular-ligament system, thus affecting on limitation of limb function and gait pathology.
Population studies indicate that approximately 35% of young patients who have undergone THA require additional follow-up visits during early rehabilitation
Fig. 4. Mean normalized GM activity [%MVIC] for detailed classification of hip joints for Load A and B condition after arthroplasty
period what is associated with GM failure. It was proved that the correct strength of the abductors re- turns one year after surgery [12]. Surgery treatment is considered to be terminated when painless gait with- out orthopedic supports is observed, forgetting that the strength of the abductors increases by only 50% in 6 months after THA and their dysfunction is recorded electromyographically even 2 years after surgery. The success of a quick return to full weighting of the oper- ated limb without crutches is criticized by many authors, due to the abductors insufficiency which are unable to stabilize the pelvis in such a short time. This failure may lead to joint instability and be the cause of early and late complications of arthroplasty. Hence the role of research on the function of muscle abductors both before and after surgery as well as extended rehabili- tation protocols focused on this motor unit [10].
The subject of many studies was muscle activity during gait. Hip abductors are relatively weak muscles of the lower limb, despite their important role in pel- vic stabilization. The maximum values of the abduc- tors moments of strength are smaller than the mo- ments of the hip flexors or extensors, although higher than the rotators muscle. Researchers analyzed both physiological gait in healthy individuals and various gait pathologies [7]. During the normal gait each mus- cle group exhibits their activity. The GM is the most active in the support phase and in addition at the end of the swing phase.
Undoubtedly, greater activity of the GM is ob- served during full weight bearing (Load A) compared to partial weight bearing with the load of raised limb (Load B). This relationship has been confirmed in the Trendelenburg test for both limbs, affected by OA and opposite, before (Fig. 1; F (1.44) = 13.8; p < 0.005) and after (Fig. 2, F (1.120) = 79.56; p < 0.0001) sur- gery. Similar relationship in healthy people was demon- strated by other authors by electromyographic analysis of 22 exercises activating gluteus muscles where the
“pelvic drop” test showed activity at the level of 58.43% MVIC which placed this exercise on the 11th position in terms of GM activity [16], [20]. It was also shown that for full weight bearing conditions there is a trend towards asymmetric muscle activity (Fig. 1), the greatest on the side with OA (mean 40.5% MVIC) in relation to the contralateral joint (mean 25% MVIC).
This asymmetry is reduced after THA (Fig. 2) and the level of activity is inverse (lower on the operated side). Although GM in hip affected by OA is ineffi- cient and certainly weaker than healthy, it generates greater electromyographic signals.
Imaging studies additionally confirm the partial atrophy of both superficial and deep muscles in the
region of affected hip [6]. In this study muscular atro- phy was not assessed, although it was expected in OA hips. Higher EMG activity of muscles around the af- fected joints was also received by other authors ana- lyzing GM in hip OA as well as quadriceps muscle in knee OA [3], [16], [21]. Several studies try to explain this phenomenon. The hypothesis that patients with hip OA activate muscle more due to faster movements is not confirmed in gait laboratory tests, because they show a significantly lower gait speed than healthy people. Definitely the need to keep the pelvis hori- zontally in the frontal plane, thus, maintaining the right center of gravity forces greater abductors’ activity during standing. The dependence of the osteoarticular system on the musculoskeletal unit is confirmed in many pathologies of the musculoskeletal system.
However, the question should be asked whether the primary problem concerns the osteoarticular system and results in muscular dysfunction, or is the opposite situation, and muscular dysfunction predispose to the development of joint disease? Our study, unfortunately, does not explain the order of events, but it undoubt- edly confirms the relationship of these two systems [15]. It is possible that the pathology of both systems coexists, and the central nervous system generates more neuromuscular connections by recruiting more motor units to strengthen muscle strength enough to make the move [14]. Advanced OA is also associated with significant pain that stimulates increased muscle tension. In this study a detailed analysis of contralat- eral muscle activity has been performed. It was shown that for full weight bearing (Load A, Figs. 3, 4), the joint left for conservative treatment or for subsequent intervention obtained the highest activity, while the joints after THA presented an activity level similar to healthy joints. Low weight bearing tests (Load B, Figs. 1–4) showed no differentiation in activity re- gardless of joint condition. Perhaps, it was reason why it was difficult to show differences in muscle activity in healthy and pathologically affected joints with other low weight bearing tests. Therefore, it seems that the proposed schematic of full (Load A) and partial (Load B) weight bearing confirms the usefulness of the Trendelenburg test in further preoperative EMG signals of GM for hips qualified for THA and in post- operative period, as well as contralateral hips in vari- ous clinical conditions. Perhaps, the time of the post- operative examination is important for these results.
Significant improvement in the general condition of patients, pain relief, physical disability reduction, as well as improvement of mood and return to social life are observed 3 months after surgery [1]. However, some authors indicate a postoperative period of six months as
the most important for comprehensive rehabilitation, although gait speed and symmetry of gait improves up to a year after surgery [9]. Our observations indicate an improvement in gait parameters 6 month after op- eration, although it did not match the values of healthy people [13]. This period allows for the restoration of proper mobility of the operated hip joint, however, improper kinematic of the opposite limb is still ob- served, especially the knee joint. The correlation and impact of joint surgery on the entire musculoskeletal system and the resulting gait pathologies are com- monly described in the literature and do not only apply to arthroplasties [5].
The GM muscle activity on the non-operated hip increases in the group with bilateral OA (CC). An increase in EMG activity could be associated with a full weight bearing of opposite limb, which was qualified for arthroplasty later. However, all exam- ined patients walked with full weight bearing without orthopedic support in postoperative assessment. This study did not include proprioceptive analysis, how- ever, because of changes in the neuromuscular activ- ity of the operated hip, abnormalities of the opposite hip proprioception and the resulting differences in EMG activity of a clinically better joint should be expected. Similar results were obtained in Dwyer’s study, which states increased GM electromyographic signals of both hips in patients with advanced unilat- eral hip OA [3].
Limitations
Firstly, due to the difficulty in obtaining homoge- neous research material, not only in terms of the hip joint condition, but also other factors characterizing the examined patients, individual hip joints were identi- fied as independent units. For this reason, the popular repeated measurement scheme cannot be used in sta- tistical analysis, since not all joints assessed before surgery were also examined after surgery and vice versa. Some people qualified for surgery were dis- qualified from the study because of anesthesia rea- sons, others did not comply with the obligation of a follow-up examination. Due to the above-mentioned conditions, it was decided to perform test statistics for independent measurements. Secondly, the small sam- ple size, especially in the CC and CE subgroup, probably does not allow to indicate other statistically significant differences confirming the trends of mean values. However, the origin of the study group from one center, the standardized method of treatment and rehabilitation gives the possibility of objective analy- sis of the studied parameters.
5. Conclusions
The analysis of the obtained results of abductors activity after hip replacement emphasizes the diversity in muscle activity resulting from the state of weight bearing, and reveals a higher bioelectric activity of abductor muscle opposite to the operated hip, affected by osteoarthritis. The obtained findings take notice of postoperative rehabilitation protocols, which should be differentiated relatively to the clinical condition of the non-operated limb. It seems reasonable to extend the rehabilitation protocol of this group of patients to non- operated limb, especially with regard to pelvic stabili- zation and efficiency of the GM muscle [12]. Increased post-operative GM electromyographic activity on the non-operated side inclines to detailed clinical control.
Prolonged waiting time for potential surgical treatment of affected, opposite hip may result in a worse func- tional outcome of abductors. There is a need for further research to explain whether unilateral OA may be the cause of the progression towards OA of opposite hip.
Acknowledgements
We would like to thank all members of the Department of Or- thopedics and Traumatology who help us to conduct the research and discuss the results. This study received funding from the Medical University of Silesia in Katowice (KNW-1-097/08 and KNW-1-108/N/9/K).
References
[1] DAVIS A.M., PERRUCCIO A.V., IBRAHIM S., HOGG-JOHNSON S., WONG R., STREINER D.L., BEATON D.E., CÔTÉ P., GIGNAC M.A., FLANNERY J., SCHEMITSCH E., MAHOMED N.N., BADLEY E.M., The trajectory of recovery and the inter-relationships of symp- toms, activity and participation in the first year following total hip and knee replacement, Osteoarthritis Cartilage, 2011, 19, 1413–1421.
[2] DE LUCA C.J., ADAM A., WOTIZ R., GILMORE L.D., NAWAB S.H., Decomposition of surface EMG signals, J. Neurophysiol., 2006, 96, 1646–1657.
[3] DWYER M.K., STAFFORD K., MATTACOLA C.G., UHL T.L., GIORDANI M., Comparison of gluteus medius muscle activity during functional tasks in individuals with and without osteo- arthritis of the hip joint, Clin. Biomech., 2013, 28, 757–761.
[4] FRENCH H.P., HUANG X., CUMMISKEY A., MELDRUM D., MALONE A., Normalisation method can affect gluteus medius electromyography results during weight bearing exercises in people with hip osteoarthritis (OA): a case control study, Gait Posture, 2015, 41, 470–475.
[5] FOUCHER K.C., WIMMER M.A., Contralateral hip and knee gait biomechanics are unchanged by total hip replacement for uni- lateral hip osteoarthritis, Gait Posture, 2012, 35, 61–65.
[6] GRIMALDI A., RICHARDSON C., STANTON W., DURBRIDGE G., DONNELLY W., HIDES J., The association between degenerative hip joint pathology and size of the gluteus medius, gluteus mini- mus and piriformis muscles, Man. Ther., 2009, 14, 605–610.
[7] HAJDUK G., NOWAK K., SOBOTA G., KUSZ D., KOPEĆ K., BŁASZCZAK E., CIELIŃSKI Ł., BACIK B., Kinematic gait parame- ters changes in patients after total knee arthroplasty. Compari- son between cruciate-retaining and posterior-substituting de- sign, Acta Bioeng. Biomech., 2016, 18, 137–142.
[8] HARDCASTLE P., NADE S., The significance of the Trendelenburg test, J. Bone Jt. Surg., 1985, 67, 741–746.
[9] HODT-BILLINGTON C., HELBOSTAD J.L., VERVAAT W., ROGNSVÅG T., MOE-NILSSEN R., Criteria of gait asymmetry in patients with hip osteoarthritis, Physiother. Theory Pract., 2012, 28, 134–141.
[10] JACOBS C.A., LEWIS M., BOLGLA L.A., CHRISTENSEN C.P., NITZ A.J., UHL T.L., Electromyographic analysis of hip ab- ductor exercises performed by a sample of total hip arthro- plasty patients, J. Arthroplasty, 2009, 24, 1130–1136.
[11] JIANG N., PENG L., AL-QWBANI M., XIE G.P., YANG Q.M., CHAI Y., ZHANG Q., YU B., Femoral version, neck-shaft angle, and acetabular anteversion in Chinese Han population:
a retrospective analysis of 466 healthy adults, Medicine (Baltimore), 2015, 94, e891.
[12] JUDD D.L., DENNIS D.A., THOMAS A.C., WOLFE P., DAYTON M.R., STEVENS-LAPSLEY J.E., Muscle strength and functional recov- ery during the first year after THA, Clin. Orthop. Relat. Res., 2014, 472, 654–664.
[13] KOPEĆ K., KUSZ D., SOBOTA G., NOWAK K., MIERZWIŃSKI M., NOWAK M., Gait analysis in patients after unilateral hip ar- throplasty, Ortop. Traumatol. Rehabil., 2015, 17, 39–50.
[14] LING S.M., CONWIT R.A., TALBOT L., SHERMACK M., WOOD J.E., DREDGE E.M., WEEKS M.J., ABERNETHY D.R., METTER E.J., Electromyographic patterns suggest changes in motor unit physiology associated with early osteoarthritis of the knee, Osteoarthritis Cartilage, 2007, 15, 1134–1140.
[15] MACADAM P., CRONIN J., CONTRERAS B., An examination of gluteal muscle activity associated with dynamic hip abduc- tion and hip external rotation exercise: a systematic review, Int. J. Sports Phys. Ther., 2015, 10, 573–591.
[16] MARSHALL A.R., NORONHA M., ZACHARIAS A., KAPAKOULAKIS T., GREEN R., Structure and function of the abductors in patients with hip osteoarthritis: Systematic review and meta-analysis, J. Back Musculoskelet. Rehabil., 2016, 29, 191–204.
[17] MARUYAMA M., FEINBERG J.R., CAPELLO W.N., D’ANTONIO J.A., The Frank Stinchfield Award: Morphologic features of the acetabulum and femur: anteversion angle and implant posi- tioning, Clin. Orthop. Relat. Res., 2001, 393, 52–65.
[18] MEEKINS G.D., SO Y., QUAN D., American Association of Neuromuscular and Electrodiagnostic Medicine evidenced- based review: use of surface electromyography in the diag- nosis and study of neuromuscular disorders, Muscle Nerve, 2008, 38, 1219–1224.
[19] PAI V.S., Significance of the Trendelenburg test in total hip arthroplasty. Influence of lateral approaches, J. Arthroplasty, 1996, 11, 174–179.
[20] REIMAN M.P., BOLGLA L.A., LOUDON J.K., A literature re- view of studies evaluating gluteus maximus and gluteus me- dius activation during rehabilitation exercises, Physiother.
Theory Pract., 2012, 28, 257–268.
[21] SIMS K.J., RICHARDSON C.A., BRAUER S.G., Investigation of hip abductor activation in subjects with clinical unilat- eral hip osteoarthritis, Ann. Rheum. Dis., 2002, 61, 687 –692.
[22] STANISZ A., Przystępny kurs statystyki z zastosowaniem STATISTICA PL na przykładach z medycyny, T. 2. Modele liniowe i nieliniowe, Statsoft Polska, 2007.
[23] YANG Y., ZUO J., LIU T., XIAO J., LIU S., GAO Z., Morpho- logical Analysis of True Acetabulum in Hip Dysplasia (Crowe Classes I–IV) Via 3-D Implantation Simulation, J. Bone Joint Surg. Am., 2017, 6, e92.
